Helicobacter pylori infection is associated with an inflammatory response in the gastric mucosa, ultimately leading to cellular hyperproliferation and malignant transformation. Hitherto, only expression of a single gene, or a limited number of genes, has been investigated in infected patients. cDNA arrays were therefore used to establish the global pattern of gene expression in gastric tissue of healthy subjects and of H. pylori-infected patients. Two main gene expression profiles were identified based on cluster analysis. The data obtained suggest a strong involvement of selected Toll-like receptors, adhesion molecules, chemokines, and ILs in the mucosal response. This pattern is clearly different from that observed using gastric epithelial cell lines infected in vitro with H. pylori. The presence of a “Helicobacter-infection signature,” i.e., a set of genes that are up-regulated in biopsies from H. pylori-infected patients, could be derived from this analysis. The genotype of the bacteria (presence of genes encoding cytotoxin-associated Ag, vacuolating cytotoxin, and blood group Ag-binding adhesin) was analyzed by PCR and shown to be associated with differential expression of a subset of genes, but not the general gene expression pattern. The expression data of the array hybridization was confirmed by quantitative real-time PCR assays. Future studies may help identify gene expression patterns predictive of complications of the infection.

Helicobacter pylori infection affects more than half of the world’s population. The clinical consequences range from asymptomatic gastritis to peptic ulceration and gastric malignancy (1, 2). The outcome of the infection may be related to differences in virulence among the bacterial strains or depend on host factors. Several virulence factors of H. pylori, including the vacuolating cytotoxin (VacA)3 and the cytotoxin-associated Ag (CagA), have been identified and their presence defines the virulent “type 1” bacteria (3) that are associated with a more severe disease status (4). The well-defined bacterial adherence factor blood group Ag-binding adhesin (BabA) (5) may also play an important role in the induction of severe gastric inflammation, particularly when associated with CagA and VacA (6). More recently, the H. pylori sialic acid-binding adhesin has also been implicated in persistent infection and chronic inflammation (7).

The nature of the host immune response may also determine the clinical outcome of the infection. H. pylori colonization induces a strong systemic and mucosal immune response (8, 9, 10). However, Ab production does not result in eradication of the infection, even though H. pylori is susceptible in vitro to Ab-dependent complement-mediated phagocytosis and killing (11). Rather, autoantibody-mediated destruction of epithelial cells may initiate or maintain mucosal inflammation (12, 13, 14).

Both CD4+ and CD8+ T cells are increased at the site of infection with H. pylori (15). The majority of CD4+ T cell clones isolated from H. pylori-infected individuals express a Th1 phenotype, and produce high levels of IFN-γ and IL-12 (15, 16, 17). This Th1 deviation may be associated with increased antral production of IL-18 in response to H. pylori infection (18). The skewed host response (Th1), combined with Fas-mediated apoptosis of H. pylori-specific T cell clones (19), may contribute to the persistence of the infection and studies in mice have suggested that a balanced Th1 and Th2 response is necessary for protection (2, 20).

Host genetic factors may also influence the immune and inflammatory response to H. pylori infection. Polymorphisms in the IL-1 gene cluster were shown to be associated with an increased risk of hypochlorhydria induced by H. pylori and gastric adenocarcinoma (21). A recent study also showed that polymorphisms in the gene encoding the IFN-γR1 are associated with susceptibility to infection (22).

Many studies of the pathogenesis of H. pylori-associated gastritis have been based on the analysis of the expression of single or a relatively limited number of cytokine or chemokine genes in the infected gastric mucosa (for review, see Ref. 23). However, the global pattern of inflammatory gene expression has not been analyzed. Recently, microarray techniques have become available that allow characterization of the mRNA expression pattern of large numbers of genes simultaneously. In this study, we used an inflammatory cDNA array, representing a comprehensive collection of genes encoding cytokines, chemokines, their receptors, and other immunomodulatory factors, to analyze the profile of gene expression in H. pylori-associated gastritis. Two main profiles of gene expression were identified based on cluster analysis. Comparison of these profiles with clinical and pathological data may not only contribute to the identification of genes associated with H. pylori infection and to a better understanding of the pathophysiological mechanisms involved, but may also help to predict the possible outcome of the infection in patients.

Gastric biopsies were collected from 21 individuals who underwent gastroscopy due to dyspepsia, upper abdominal pain, or a routine check before partial gastrectomy (obesity patients). Patients with malignancy, immunosuppression, metabolic disorders, gastrointestinal hemorrhage, or who were receiving aspirin or other nonsteroidal anti-inflammatory drugs were excluded from the study. The patients are all central-European residents, all belonging to the Caucasian population except one recent immigrant from Asia (patient 64). Multiple endoscopic biopsies were obtained from the antrum and the corpus for parallel assessment of gene expression and histology. For RNA preparation, biopsies were flash-frozen in liquid nitrogen and stored at −70°C. Table I summarizes the characteristics of the patients. The Institution Review Boards at the University Hospital (Lausanne, Switzerland) and the Karolinska Institute (Stockholm, Sweden) approved the study, and informed consent was obtained from all patients.

Table I.

Characterization of patients and grading of gastritisa

CodeAgeSexClinical DiagnosisHistology Results
Presence of H. pyloriNeutrophil infiltrationMononuclear cell infiltrationAtrophyIntestinal metaplasia
H. pylori-negative group (n = 9)         
45 43 Severe obesity Absent Absent Absent Absent Absent 
23 38 Severe obesity Absent Absent Absent Absent Absent 
24 42 Severe obesity Absent Absent Absent Absent Absent 
13 48 Gastroreflux disease Absent Absent Absent Absent Absent 
28 35 Moderate obesity Absent Absent Absent Absent Absent 
39 Gastroreflux disease, successful eradicationb Absent Absent Mild Absent Absent 
54 38 Epigastric pain, successful eradicationb Absent Absent Mild Absent Absent 
44 Dyspepsia, chronic epigastric pain, successful eradicationb Absent Absent Mild Absent Absent 
50 50 Dyspepsia, chronic epigastric pain Absent Absent Mild Absent Absent 
H. pylori-positive group (n = 12)         
65 Gastritis, unsuccessful eradication Marked Mild Mild Absent Absent 
49 18 Epigastric pain Mild Mild Moderate Absent Present 
81 33 Gastritis, unsuccessful eradication Moderate Mild Moderate Absent Absent 
12 33 Gastritis, unsuccessful eradication Marked Mild Moderate Absent Absent 
22 29 Gastritis Marked Mild Moderate Absent Absent 
87 25 Gastritis, unsuccessful eradication Marked Mild Moderate Present Absent 
27 26 Gastritis Moderate Moderate Moderate Absent Absent 
64 32 Gastritis Marked Moderate Moderate Absent Absent 
11 38 Gastroreflux disease, unsuccessful eradication Marked Moderate Moderate Absent Absent 
30 Gastritis, unsuccessful eradication Marked Mild Marked Present Present 
50 Gastritis, unsuccessful eradication Moderate Mild Marked Present Absent 
65 55 Epigastric pain, gastric ulcer Moderate Mild Marked Absent Absent 
CodeAgeSexClinical DiagnosisHistology Results
Presence of H. pyloriNeutrophil infiltrationMononuclear cell infiltrationAtrophyIntestinal metaplasia
H. pylori-negative group (n = 9)         
45 43 Severe obesity Absent Absent Absent Absent Absent 
23 38 Severe obesity Absent Absent Absent Absent Absent 
24 42 Severe obesity Absent Absent Absent Absent Absent 
13 48 Gastroreflux disease Absent Absent Absent Absent Absent 
28 35 Moderate obesity Absent Absent Absent Absent Absent 
39 Gastroreflux disease, successful eradicationb Absent Absent Mild Absent Absent 
54 38 Epigastric pain, successful eradicationb Absent Absent Mild Absent Absent 
44 Dyspepsia, chronic epigastric pain, successful eradicationb Absent Absent Mild Absent Absent 
50 50 Dyspepsia, chronic epigastric pain Absent Absent Mild Absent Absent 
H. pylori-positive group (n = 12)         
65 Gastritis, unsuccessful eradication Marked Mild Mild Absent Absent 
49 18 Epigastric pain Mild Mild Moderate Absent Present 
81 33 Gastritis, unsuccessful eradication Moderate Mild Moderate Absent Absent 
12 33 Gastritis, unsuccessful eradication Marked Mild Moderate Absent Absent 
22 29 Gastritis Marked Mild Moderate Absent Absent 
87 25 Gastritis, unsuccessful eradication Marked Mild Moderate Present Absent 
27 26 Gastritis Moderate Moderate Moderate Absent Absent 
64 32 Gastritis Marked Moderate Moderate Absent Absent 
11 38 Gastroreflux disease, unsuccessful eradication Marked Moderate Moderate Absent Absent 
30 Gastritis, unsuccessful eradication Marked Mild Marked Present Present 
50 Gastritis, unsuccessful eradication Moderate Mild Marked Present Absent 
65 55 Epigastric pain, gastric ulcer Moderate Mild Marked Absent Absent 
a

The degree of inflammation present in antrum biopsies were classified according to the updated Sydney system (24 ). A grading from absent to marked was assigned for four histological variables: chronic inflammation (mononuclear cell infiltration), activity (neutrophil infiltration), glandular atrophy, and intestinal metaplasia.

b

Successful eradication of H. pylori infection.

Four gastric biopsies (two from antrum and two from corpus) were obtained from each patient, fixed in formalin, and subsequently evaluated by an experienced pathologist, blinded to all other aspects of the study. The degree of inflammation present in the histological specimens was classified according to the updated Sydney system (24). A grading from absent to marked was assigned for four histological variables: chronic inflammation (mononuclear cell infiltration), activity (polymorphonuclear neutrophil infiltration), glandular atrophy, and intestinal metaplasia. The presence of H. pylori infection was determined either by visualization by histology (including Giemsa staining) and/or a positive rapid urease test performed on at least one additional biopsy sample.

The human cytokine expression array (R&D Systems, Abingdon, U.K.), which consists of 847 different cloned cDNAs representing a comprehensive collection of cytokines, chemokines and other immunomodulatory factors, and their receptors (www.rndsystems.com or www.ncbi.nlm.nih.gov/geo/, platform GPL193), was used for the study. Nine positive control “housekeeping” genes, six negative controls, and human genomic DNA are also included in the array.

Total RNA was isolated from ∼20 to 30 mg of frozen gastric biopsy samples (two biopsies) using the RNeasy mini kit (Qiagen, Hilden, Germany). RNA samples from antrum and corpus biopsies were prepared separately. In addition to the standard procedure, DNase I (Qiagen) was used to remove trace amounts of genomic DNA. Total RNA was quantified by measuring the absorbance at 260 and 280 nm (A260 and A280) and the integrity was assessed by agarose gel electrophoresis before radiolabeling. cDNA probe preparation and membrane hybridization were performed according to the respective kit manuals (R&D Systems). Briefly, 2 μg of total RNA were reverse transcribed into cDNA by avian myeloblastosis virus reverse transcriptase using a human cytokine-specific primer mix in the presence of [α-33P]dCTP (2000 Ci/mmol; Amersham Pharmacia, Uppsala, Sweden) and 0.3 mM of a dATP/dGTP/dTTP mixture at 42°C for 2 h. Labeled cDNA was subsequently separated from unincorporated nucleotides using Sephadex G-25 spin columns (R&D Systems). The array membranes were prehybridized at 65°C for 2 h in hybridization solution (5× SSPE, 2% (w/v) SDS, 5× Denhardt’s reagent, and 100 μg/ml sonicated, salmon DNA). Labeled probes were added to 3 ml of fresh hybridization solution and allowed to hybridize to the array membranes at 65°C overnight. After hybridization, membranes were washed two times with solution 1 (0.5× SSC and 1% SDS), twice with solution 2 (0.1× SSC and 1% SDS) for 20 min at 65°C, and finally exposed to a phosphorus screen for 7 days. Each array was not reused more than one time after stripping.

The cDNA microarrays were scanned and exported as image and information files using a Fuji Bio-image Analyzer BAS2000 (Fuji, Düsseldorf, Germany). The images and quantitative data of gene expression levels were analyzed using the Fuji Image Reader FLA-3000/3000G software. After the quantitative data had been recorded, each spot was also manually examined to avoid artifacts.

Spreadsheets exported from the image analysis software were analyzed using Microsoft Excel (Redmond, WA). A mean signal intensity value of 92 evenly distributed empty spots (with no cDNA spotted) and the six negative control spots of each array were used as a background value, which was subsequently subtracted from the signal intensities. The background value was used to establish the signal threshold to filter out weak signals that were not attributable to actual gene expression or those that were too weak for meaningful interpretation. The threshold was set such that any gene whose adjusted intensity was at least 2 SD above the background value was considered to be a genuine signal. To compare the results of different hybridization experiments, the signal intensity of each gene from different arrays was normalized by the averaged signal from nine housekeeping genes in each array. The corresponding normalized signals on different arrays were then compared in order to determine fold-induction or fold-reduction in expression of gene-specific RNA between samples. Only relative changes equal to, or greater than, 2-fold levels of gene expression were considered to represent up- or down-regulation. In addition, if the signal intensity of a given gene was changed from negative (below the signal threshold) to positive or from positive to negative, the gene was considered being up- or down-regulated, respectively. The complete array data are available from gene expression omnibus (accession numbers: GSM8905-8925 and GSM12762-12767)

The normalized gene expression levels from different sample groups (H. pylori-positive or -negative, cagA-positive or -negative, babA2-positive or -negative) were also compared by Mann-Whitney test using SPSS software (Chicago, IL). Multidimensional scaling (using the additive (“Manhattan”) metric) and the standard Student’s t test with multiple testing correction were performed using software R (www.R-project.org).

Hierarchical clustering was performed by using the CLUSTER program (25) and the results were displayed using TREEVIEW (software available at http://genome-www4.stanford.edu/MicroArray/SMD/restech.html).

A series of primers (Table II) were designed for PCR amplification of the 16S rRNA, cagA, vacA1, vacA2, and babA2 genes directly from the H. pylori-infected gastric biopsies. First-strand cDNA synthesis was performed with a pd(N)6 primer using a cDNA synthesis kit (Amersham Pharmacia). For cDNA synthesis, 3 μg of total RNA were used. Amplification was performed in 40 cycles, each cycle consisting of 94°C 30 s, 55°C (16S rRNA) or 60°C (cagA, vacA1, vacA2, and babA2) 30 s and 72°C 30 s. The PCR products were analyzed on a 2% agarose gel stained with ethidium bromide and their identities were confirmed by cloning and sequencing of selected samples.

Table II.

Primers used for detecting H. pylori-specific genesa

GenePrimerProduct Size (bp)
16S rRNA 5′-GGAGGCAGCAGTAGGGAATA-3′ 392 
 5′-GCAATCAGCGTCAGTAATGT-3′  
babA2 5′-AATCCAAAAAGGAGAAAAAACATGAAA-3′ 488 
 5′-CTGCAAGTGATGGAAGTGGAT-3′  
cagA 5′-TTAGACAACTTGAGCGAGAAAGA-3′ 566 
 5′-GCKbTCAGCTACAGCTTTATTGAA-3′  
vacA 5′-ATGGAAATACAACAAACACAC-3′ 259 (s1) 
s1/s2 5′-CTGCTTGAATGCGCCAAAC-3′ 286 (s2) 
GenePrimerProduct Size (bp)
16S rRNA 5′-GGAGGCAGCAGTAGGGAATA-3′ 392 
 5′-GCAATCAGCGTCAGTAATGT-3′  
babA2 5′-AATCCAAAAAGGAGAAAAAACATGAAA-3′ 488 
 5′-CTGCAAGTGATGGAAGTGGAT-3′  
cagA 5′-TTAGACAACTTGAGCGAGAAAGA-3′ 566 
 5′-GCKbTCAGCTACAGCTTTATTGAA-3′  
vacA 5′-ATGGAAATACAACAAACACAC-3′ 259 (s1) 
s1/s2 5′-CTGCTTGAATGCGCCAAAC-3′ 286 (s2) 
a

Primers for 16SrRNA, babA2, and cagA were designed based on the published sequences for the respective genes (U00679, AF033654, and AF247651) and primers for vacA gene were described previously (58 ).

b

K, G, or T.

First-strand cDNA synthesis was performed with a NotI-d(T)18 primer using a cDNA synthesis kit (Amersham Pharmacia). For cDNA synthesis, 2.35 μg of total RNA was used. Primer sequences are shown in supplementary data Table I.4 The amplification was performed using the qPCR core kit for SYBR Green I (MedProbe, Oslo, Norway) and the ABI PRISM 7000 sequence detection system (Applied Biosystems, Foster City, CA). Typical profile times used were: initial step, 95°C, 10 min followed by a second step, 95°C 15 s and 60°C 1 min, 40 cycles. The standard curves for a housekeeping gene (GAPDH) and the target gene were generated by serial dilutions of the control cDNA sample (equivalent to 80 ng of total RNA) in five 2-fold dilution steps and used for regression analysis. The amount of target and housekeeping gene in every sample was determined from the respective standard curves and the target amount was divided by the housekeeping gene expression level to obtain a normalized target value (relative expression level).

In a first set of experiments, array hybridization using different conditions (varying amounts of RNA input, different exposure time etc.) was performed to optimize the test protocols. α-32P vs α-33P labeling was also tested and as α-33P gave a lower background, it was chosen for all subsequent experiments.

Twelve H. pylori-positive and nine H. pylori-negative antrum biopsies were subsequently tested in parallel on inflammatory cDNA arrays. On average, expression of 78.5% of the genes on the arrays could be detected and the number of detectable genes in the H. pylori-infected group was significantly higher than in the noninfected group (83.9 and 71.3%, respectively, t test, p < 0.01). To verify the reproducibility, four patient samples were tested twice and showed a >95% correlation.

The gene expression patterns, as evaluated from the hybridization images, were clearly different between the H. pylori-infected and noninfected groups and a number of genes were differentially expressed between the two groups. A typical example is given in Fig. 1 (labeled A–Z).

FIGURE 1.

Hybridization images for one H. pylori-negative (A) and one positive (B) sample. Selected genes that exhibit different expression patterns are marked with arrows and letters.

FIGURE 1.

Hybridization images for one H. pylori-negative (A) and one positive (B) sample. Selected genes that exhibit different expression patterns are marked with arrows and letters.

Close modal

To characterize the overall patterns of gene expression and group the samples accordingly, a hierarchical clustering algorithm was subsequently applied. The result is presented graphically as colored images (Fig. 2). Along the vertical axis, the genes analyzed are arranged as ordered by the clustering algorithm, where the genes with the most similar patterns of expression are placed adjacent to each other. Along the horizontal axis, the samples are arranged such that those with the most similar patterns of expression across all genes are placed adjacent to each other. Two main gene expression profiles could be identified based on the cluster analysis and the samples were grouped accordingly (Fig. 2). Group I was shown to encompass all the H. pylori-positive samples whereas group II contained the H. pylori-negative samples.

FIGURE 2.

Cluster diagram of gene expression in gastric biopsies. Each column represents one biopsy and each row represents a single gene. The raw data from the array experiments were first normalized as described in Materials and Methods and then log transformed before the cluster analysis. Mean level of gene expression in each array experiment was calculated and the expression level of each gene relative to the mean level is shown. Green squares represent lower than mean levels of gene expression in the individual biopsy samples; black squares represent genes with mean levels of expression; red squares represent higher than mean levels of gene expression; gray squares indicate insufficient or missing data. The color saturation reflects the magnitude of the log/ratio. A, Overview of gene expression in all the samples. B, Expanded view of the gene clusters that constitute the H. pylori-infection signature. ∗, Genes that were differentially expressed between the H. pylori positive and negative samples, using stringent criteria as described in Results. #, Genes that were differentially expressed between the babA2-positive and -negative samples. IL-1RL2, IL-1R-like 2.

FIGURE 2.

Cluster diagram of gene expression in gastric biopsies. Each column represents one biopsy and each row represents a single gene. The raw data from the array experiments were first normalized as described in Materials and Methods and then log transformed before the cluster analysis. Mean level of gene expression in each array experiment was calculated and the expression level of each gene relative to the mean level is shown. Green squares represent lower than mean levels of gene expression in the individual biopsy samples; black squares represent genes with mean levels of expression; red squares represent higher than mean levels of gene expression; gray squares indicate insufficient or missing data. The color saturation reflects the magnitude of the log/ratio. A, Overview of gene expression in all the samples. B, Expanded view of the gene clusters that constitute the H. pylori-infection signature. ∗, Genes that were differentially expressed between the H. pylori positive and negative samples, using stringent criteria as described in Results. #, Genes that were differentially expressed between the babA2-positive and -negative samples. IL-1RL2, IL-1R-like 2.

Close modal

Several clusters of genes were identified that could serve as signatures for H. pylori infection (Fig. 2,B and for a larger version, see supplementary Fig. 1). These include genes encoding the IL-1R/Toll-like receptor (TLR) family (IL-18 receptor accessory protein (RAP), TLR1, radioprotective 105-kDa protein (RP105), TLR6, TLR5, TLR4), cytokine and cytokine receptors (IL-2Rβ, TNF-α, IL-10Rα, IL-16, IFN-γR1, GM-CSFRβ, IL-2Rγ), complement and complement receptors (CR1, CD21, C3), adhesion molecules (ICAM-1, VCAM-1, mucosal vascular addressin cell adhesion molecule 1 (MadCAM-1)), integrins (integrin-β2, integrin-αX, integrin-αL, integrin-α4, integrin-β7), chemokines and chemokine receptors (B lymphocyte chemoattractant (BLC)/B-lymphocyte chemoattractant (BCA)-1, macrophage inflammatory protein (MIP)-3α, epithelia-derived neutrophil-activating peptide (ENA)-78, growth-related protein (GRO)-β, GRO-α, GRO-γ, CCR-7, CCR-6, CXCR-4), cell signaling molecules (mitogen-activated protein kinase (MAPK)10, DAP kinase-related apoptosis-inducing protein kinase (DRAK)2, protein kinase C β1 (PRKCB1)), proteinases (matrix metalloproteinase (MMP)-9, a disintegrin and metalloproteinase (ADAM)12), and some factors that are important for B or T cell activation (CD45, CD53, GL50/B7-H2, signal lymphocytic activation molecule (SLAM), B7-H1).

To further identify the genes that were differentially expressed between the two groups, we performed an analysis using stringent criteria (requiring consistency in the expression values across a majority of the samples) where the normalized signal of each individual gene in a given sample in the H. pylori-positive group (n = 12) was compared with the data of the corresponding gene in each of the control samples in the H. pylori-negative group (n = 9). If the signal intensity was higher than in the control sample, i.e., showing at least a 2-fold increase or a change from being undetectable to detectable, in >80% of these comparisons (n = 108), the gene was considered being up-regulated. Conversely, if the expression was lower than in the H. pylori-negative samples, it was considered being down-regulated. As shown in Table III, 37 genes were up-regulated in the H. pylori-infected samples. All these genes were actually present among the “H. pylori-infection signature” identified by the cluster analysis (Fig. 2,B, marked by ∗). We next performed a statistical analysis (Mann-Whitney test) and altogether 232 genes were shown to be differentially expressed in the H. pylori-positive and -negative samples (55 genes, p < 0.001; 86 genes, 0.001 ≤ p < 0.01; 91 genes, 0.01 ≤ p < 0.05, for details see supplementary data, Table II). Using this test, all the genes in the H. pylori-infection signature, identified by the cluster analysis, were differentially expressed to a statistically significant degree (p < 0.01) and the 37 genes identified by the stringent criteria (Table III) were among those with the highest significance (p ≤ 0.001). The Student t test with multiple testing corrections was also performed and the list of genes with high significance largely overlapped with that shown by the Mann-Whitney test (data not shown).

Table III.

Differentially expressed genes in H. pylori-infected gastric mucosa

Gene NameHP+/HP (n = 12/n = 9)Normalized Expression Level (Mean ± SD) (HP+/HP)Relative Fold InductionbGenbank Accession No.Cellular/Tissue Expressiona
Adhesion molecules      
VCAM-1 ↑ 2.6 ± 1.1/0.1 ± 0.2 24.8 NM_001078 Endo, D 
MadCAM-1 ↑ 3.8 ± 2.0/0.9 ± 0.8 4.3 NM_007164 Endo 
Apoptosis-related proteins      
ALOX5 ↑ 3.0 ± 1.0/0.9 ± 0.4 3.5 NM_000698 Macro 
Binding proteins      
BPI ↑ 14.0 ± 11.5/1.0 ± 1.1 13.9 NM_001725 Epi 
      
Cell surface proteins      
TLR5 ↑ 6.3 ± 4.8/0.7 ± 0.6 8.6 AF051151 Mono 
B7-H1 ↑ 51.2 ± 41.7/4.7 ± 4.3 11.0 NM_014143 Endo, D 
CD53 ↑ 5.5 ± 2.6/1.4 ± 0.9 4.0 NM_000560 B, Mono, Macro, Neutro, T 
TLR6 ↑ 4.1 ± 3.0/0.4 ± 0.6 9.4 NM_006068 Mono, immature D, Endo 
SLAM ↑ 32.6 ± 25.5/3.8 ± 2.5 8.6 NM_003037 Memory T, T, immature B 
C3 ↑ 30.0 ± 14.3/1.8 ± 0.8 16.2 NM_000064 Mono, Macro 
CD45/Ly5 ↑ 7.8 ± 3.0/2.0 ± 0.8 4.0 NM_002838 Hemopoietic cells 
TOSO ↑ 4.1 ± 2.4/0.1 ± 0.2 54.9 NM_005449 
RP105 ↑ 8.7 ± 6.3/0.4 ± 0.5 23.7 NM_005582 
TLR4 ↑ 3.3 ± 2.4/0.5 ± 0.4 7.1 NM_003266 Mono, Macro, D, T 
CD21 (CR2) ↑ 4.2 ± 4.2/0.1 ± 0.3 30.7 NM_001877 Mature B, T, D, Epi 
TLR1 ↑ 7.4 ± 5.1/1.3 ± 0.8 5.8 NM_003263 Ubiquitous 
Chemokines and chemokine receptors      
GRO-γ ↑ 3.4 ± 2.0/0.4 ± 0.7 8.1 NM_002090 Epi, Endo, Mono, Macro 
BLC/BCA-1 ↑ 2.4 ± 2.5/0 ± 0 c NM_006419 
ENA-78 ↑ 6.1 ± 5.3/1.0 ± 2.5 6.3 NM_002994 Epi, Mono 
GRO-β ↑ 3.0 ± 1.4/0.3 ± 0.9 10.2 NM_002089 Epi, Endo, Mono, Macro 
Cytokines and cytokine receptors      
IFN-γR1 ↑ 12.3 ± 7.1/2.5 ± 1.0 4.9 NM_000416 Macro, Mono, B, Endo 
MER ↑ 23.6 ± 18.4/3.4 ± 1.7 7.0 NM_006343 Neoplastic B and T cell lines 
ILs and IL receptors      
GM-CSFRβ ↑ 6.3 ± 2.8/1.6 ± 0.9 4.0 NM_000395 Myeloid progenitors, Granu 
IL-10Rα ↑ 2.4 ± 1.0/0.4 ± 0.5 6.4 NM_001558 B, Th, Mono, Macro, Epi 
IL-18RAP ↑ 4.3 ± 4.1/0.1 ± 0.2 37.7 NM_003853 PBL 
Integrins      
αX ↑ 2.8 ± 1.4/0.2 ± 0.5 11.8 NM_000887 Myeloid cells 
β2 ↑ 5.0 ± 2.4/1.1 ± 0.6 4.4 NM_000211 Leukocytes 
Other factors      
NMA ↑ 15.2 ± 11.6/1.4 ± 1.3 10.7 NM_012342 Kidney, placenta, spleen 
LTF ↑ 94.5 ± 47/4.8 ± 2.5 19.9 NM_002343 BM, spleen, lung 
GL50/B7RP1 (B7-H2) ↑ 101.6 ± 71.8/14.7 ± 10.4 6.9 AF199028 Epi, B, Mono, D 
Proteases or related factors      
ADAM-12 ↑ 10.1 ± 7.9/0.9 ± 0.9 11.6 NM_003474 Muscle, lung 
Signal transduction factors      
MAPK10 ↑ 4.8 ± 3.1/0.7 ± 0.3 6.6 NM_002753 Brain, pancreas 
PRKCB1 ↑ 18.9 ± 14.7/2.0 ± 1.3 9.7 NM_002738 Brain, BM, spleen 
TGF-β superfamily proteins      
AMHR2 ↑ 51.5 ± 43.7/4.1 ± 3.7 12.7 NM_020547 Spleen 
TNF superfamily proteins      
EDAR ↑ 16.1 ± 13.0/1.5 ± 1.3 11.0 NM_022336 Prostate, kidney, skin 
TNFRSF7 (CD27) ↑ 4.0 ± 2.1/0.8 ± 0.8 4.8 NM_001242 Thy, T, NK, B 
TNFSF3 (LT-βR) ↑ 4.9 ± 2.6/1.1 ± 0.6 4.5 NM_002341 Mature T, B, NK 
Total number of genes ↑ 37     
Gene NameHP+/HP (n = 12/n = 9)Normalized Expression Level (Mean ± SD) (HP+/HP)Relative Fold InductionbGenbank Accession No.Cellular/Tissue Expressiona
Adhesion molecules      
VCAM-1 ↑ 2.6 ± 1.1/0.1 ± 0.2 24.8 NM_001078 Endo, D 
MadCAM-1 ↑ 3.8 ± 2.0/0.9 ± 0.8 4.3 NM_007164 Endo 
Apoptosis-related proteins      
ALOX5 ↑ 3.0 ± 1.0/0.9 ± 0.4 3.5 NM_000698 Macro 
Binding proteins      
BPI ↑ 14.0 ± 11.5/1.0 ± 1.1 13.9 NM_001725 Epi 
      
Cell surface proteins      
TLR5 ↑ 6.3 ± 4.8/0.7 ± 0.6 8.6 AF051151 Mono 
B7-H1 ↑ 51.2 ± 41.7/4.7 ± 4.3 11.0 NM_014143 Endo, D 
CD53 ↑ 5.5 ± 2.6/1.4 ± 0.9 4.0 NM_000560 B, Mono, Macro, Neutro, T 
TLR6 ↑ 4.1 ± 3.0/0.4 ± 0.6 9.4 NM_006068 Mono, immature D, Endo 
SLAM ↑ 32.6 ± 25.5/3.8 ± 2.5 8.6 NM_003037 Memory T, T, immature B 
C3 ↑ 30.0 ± 14.3/1.8 ± 0.8 16.2 NM_000064 Mono, Macro 
CD45/Ly5 ↑ 7.8 ± 3.0/2.0 ± 0.8 4.0 NM_002838 Hemopoietic cells 
TOSO ↑ 4.1 ± 2.4/0.1 ± 0.2 54.9 NM_005449 
RP105 ↑ 8.7 ± 6.3/0.4 ± 0.5 23.7 NM_005582 
TLR4 ↑ 3.3 ± 2.4/0.5 ± 0.4 7.1 NM_003266 Mono, Macro, D, T 
CD21 (CR2) ↑ 4.2 ± 4.2/0.1 ± 0.3 30.7 NM_001877 Mature B, T, D, Epi 
TLR1 ↑ 7.4 ± 5.1/1.3 ± 0.8 5.8 NM_003263 Ubiquitous 
Chemokines and chemokine receptors      
GRO-γ ↑ 3.4 ± 2.0/0.4 ± 0.7 8.1 NM_002090 Epi, Endo, Mono, Macro 
BLC/BCA-1 ↑ 2.4 ± 2.5/0 ± 0 c NM_006419 
ENA-78 ↑ 6.1 ± 5.3/1.0 ± 2.5 6.3 NM_002994 Epi, Mono 
GRO-β ↑ 3.0 ± 1.4/0.3 ± 0.9 10.2 NM_002089 Epi, Endo, Mono, Macro 
Cytokines and cytokine receptors      
IFN-γR1 ↑ 12.3 ± 7.1/2.5 ± 1.0 4.9 NM_000416 Macro, Mono, B, Endo 
MER ↑ 23.6 ± 18.4/3.4 ± 1.7 7.0 NM_006343 Neoplastic B and T cell lines 
ILs and IL receptors      
GM-CSFRβ ↑ 6.3 ± 2.8/1.6 ± 0.9 4.0 NM_000395 Myeloid progenitors, Granu 
IL-10Rα ↑ 2.4 ± 1.0/0.4 ± 0.5 6.4 NM_001558 B, Th, Mono, Macro, Epi 
IL-18RAP ↑ 4.3 ± 4.1/0.1 ± 0.2 37.7 NM_003853 PBL 
Integrins      
αX ↑ 2.8 ± 1.4/0.2 ± 0.5 11.8 NM_000887 Myeloid cells 
β2 ↑ 5.0 ± 2.4/1.1 ± 0.6 4.4 NM_000211 Leukocytes 
Other factors      
NMA ↑ 15.2 ± 11.6/1.4 ± 1.3 10.7 NM_012342 Kidney, placenta, spleen 
LTF ↑ 94.5 ± 47/4.8 ± 2.5 19.9 NM_002343 BM, spleen, lung 
GL50/B7RP1 (B7-H2) ↑ 101.6 ± 71.8/14.7 ± 10.4 6.9 AF199028 Epi, B, Mono, D 
Proteases or related factors      
ADAM-12 ↑ 10.1 ± 7.9/0.9 ± 0.9 11.6 NM_003474 Muscle, lung 
Signal transduction factors      
MAPK10 ↑ 4.8 ± 3.1/0.7 ± 0.3 6.6 NM_002753 Brain, pancreas 
PRKCB1 ↑ 18.9 ± 14.7/2.0 ± 1.3 9.7 NM_002738 Brain, BM, spleen 
TGF-β superfamily proteins      
AMHR2 ↑ 51.5 ± 43.7/4.1 ± 3.7 12.7 NM_020547 Spleen 
TNF superfamily proteins      
EDAR ↑ 16.1 ± 13.0/1.5 ± 1.3 11.0 NM_022336 Prostate, kidney, skin 
TNFRSF7 (CD27) ↑ 4.0 ± 2.1/0.8 ± 0.8 4.8 NM_001242 Thy, T, NK, B 
TNFSF3 (LT-βR) ↑ 4.9 ± 2.6/1.1 ± 0.6 4.5 NM_002341 Mature T, B, NK 
Total number of genes ↑ 37     
a

The known cellular expression of the genes included in this table was summarized from Appendix II, Immunobiology, Fifth Edition, Cytokine Reference (http://apresslp.gvpi.net), OMIM (www.ncbi.nlm.nih.gov), and when such information was not available, the potential gene expression in normal human tissue has been indicated (summarized from http://bioinfo.weizmann.ac.il/cards-bin). Endo, endothelial cells; D, dendritic cells; Macro, macrophages; Epi, epithelial cells; Mono, monocytes; B, B lymphocytes; Neutro, neutrophils; T, T lymphocytes; Granu, granulocytes; BM, bone marrow; Thy, thymocytes; Erythro, erythrocytes; Eosino, eosinophils; F, fibroblasts; P, platelet; ALOX5, arachidonate 5-lipoxygenase; MER, c-mer proto-oncogene tyrosine kinase; EDAR, ectodysplasin 1, anhidrotic receptor.

b

Relative fold induction = the average of the normalized gene expression in the H. pylori-positive group divided by the average of the normalized gene expression in the H. pylori-negative group.

c

The normalized expression levels of all the H. pylori-negative samples were zero.

Known cellular or tissue expression of the 37 up-regulated genes are summarized in Table III. The induction of some genes are clearly related to certain cell types, where bactericidal/permeability increasing protein (BPI) and MadCAM-1 are produced by epithelial and endothelial cells, respectively, whereas TOSO, RP105, BLC, TLR5 are mainly expressed in T, B, dendritic cells, and monocytes, respectively. The three CXC chemokine genes (GRO-γ, ENA-78, and GRO-β) can be produced by various cells including epithelial cells, endothelial cells, and monocytes. The specific gene expression profile in H. pylori-infected gastric mucosa is therefore most likely due to responses from many cell types, including gastric epithelial cells, endothelial cells from blood vessels, and the infiltrating immune cells.

As four of the H. pylori-negative samples exhibited mild chronic inflammation in the gastric mucosa, possibly due to a previous infection, we also compared all the H. pylori-positive samples (n = 12) with the H. pylori-negative samples that showed a normal histology and with no history of previous infection (n = 5, sample 45, 23, 24, 13, and 28). In addition to the 37 genes identified above, LPS-binding protein (LBP), platelet-derived endothelial cell growth factor (PD-ECGF), IL-2Rγ, IL-1R-like 2, β-site APP-cleaving enzyme, and DRAK2 were found to be up-regulated in the H. pylori-positive group. These genes were also present among the H. pylori-infection signature identified by the cluster analysis.

To confirm the expression data of the array hybridization, 10 genes in the H. pylori-infection signature were analyzed by quantitative real-time PCR. There was a strong correlation between the expression data from arrays and real-time PCR, where the expression of all the genes tested except MAPK10 (1.5-fold) were up-regulated ≥2-fold (Table IV).

Table IV.

Real-time PCR analysis of expression of for selected H. pylori-signature genes

Gene NameH. pylori-Positive Samples (n = 9)aH. pylori-Negative Samples (n = 4)bFold Induction (HP+/HP)
Mean CTcRelative expressiondMean CTRelative expression
ENA-78 20.06 436.31 27.01 0.12 3517.9 
C3 20.09 19.29 24.36 0.15 131.7 
LTF 17.37 12.34 21.44 0.63 19.4 
CD27 25.90 6.50 29.71 0.51 12.8 
IL-10Rα 26.75 6.41 29.19 0.78 8.2 
ADAM-12 26.17 0.68 28.75 0.09 7.4 
TLR6 22.02 3.35 23.51 0.77 4.4 
SLAM 27.16 2.84 29.42 0.66 4.3 
B7-H1 25.00 5.88 25.71 2.57 2.3 
IFN-γR1 20.89 3.68 21.51 1.87 2.0 
MAPK10 25.50 1.40 25.76 0.94 1.5 
Gene NameH. pylori-Positive Samples (n = 9)aH. pylori-Negative Samples (n = 4)bFold Induction (HP+/HP)
Mean CTcRelative expressiondMean CTRelative expression
ENA-78 20.06 436.31 27.01 0.12 3517.9 
C3 20.09 19.29 24.36 0.15 131.7 
LTF 17.37 12.34 21.44 0.63 19.4 
CD27 25.90 6.50 29.71 0.51 12.8 
IL-10Rα 26.75 6.41 29.19 0.78 8.2 
ADAM-12 26.17 0.68 28.75 0.09 7.4 
TLR6 22.02 3.35 23.51 0.77 4.4 
SLAM 27.16 2.84 29.42 0.66 4.3 
B7-H1 25.00 5.88 25.71 2.57 2.3 
IFN-γR1 20.89 3.68 21.51 1.87 2.0 
MAPK10 25.50 1.40 25.76 0.94 1.5 
a

The tested H. pylori-positive samples were: 1, 3, 22, 27, 49, 81, 87, 64, and 65.

b

The tested H. pylori-negative samples were: 8, 13, 45, and 50.

c

Threshold cycle (CT) value was normalized by the housekeeping gene GAPDH.

d

The relative expression level was calculated as described in Materials and Methods.

Three H. pylori-positive (9C, 12C, and 81C) and three H. pylori-negative (13C, 23C, and 45C) corpus biopsies were also tested in parallel on the inflammatory cDNA arrays. Using the stringent criteria described above, 6 and 3 genes were up- or down-regulated in the H. pylori-infected samples. All the up-regulated genes (BPI, C3, NMA, lactotransferrin (LTF), PRKCB1, and anti-Mullerian hormone receptor (AMHR)2) are present in the H. pylori-infection signature identified by analyzing the antrum samples (Table III and Fig. 2), suggesting that similar immune/inflammatory responses might be induced in the corpus mucosa. However, some of the responses seem to be more prominent in H. pylori-infected antrum mucosa, including VCAM-1, ENA-78, GRO-β, GRO-α, and GRO-γ, as these genes all showed a >2-fold higher expression in the H. pylori-infected antrum vs corpus samples from the same individual (n = 3). We also identified 17 genes that were differentially expressed in all six pairs of antrum/corpus samples (supplementary data, Table III). These genes, which include activated leukocyte cell adhesion molecule, cyclin-dependent kinase 6, insulin-like growth factor binding protein 2, c-kit, and leukemia inhibitory factor R, do not overlap with any of the H. pylori-infection signature; they may represent intrinsic differences between antrum and corpus.

To search for genes that are related to the degree of the chronic inflammation in the H. pylori-infected gastric samples, we divided the H. pylori-positive antrum samples into three groups based on their histology scores for chronic inflammation (marked, moderate, or mild), and compared these groups to a group of H. pylori-negative samples with a normal histology (n = 5, samples 13, 23, 24, 28, and 45) using the stringent criteria described above.

The severe chronic gastritis group included samples 1, 3, and 65 and showed a marked infiltration of mononuclear cells in the biopsies. Totally, 66 and 1 genes were being up- or down-regulated in this group as compared with the H. pylori-negative samples (Table V). The moderate gastritis group included samples 11, 12, 22, 27, 49, 64, 81, and 87 and showed a moderate infiltration of mononuclear cells. There were 47 genes up-regulated in this group, 46 of which were already identified in the severe gastritis group (Table V). The mild gastritis group included only one sample (sample 9) and the number of up-regulated genes in this sample was significantly less (n = 7) as compared with the other two groups. All of these genes were already identified in the severe and moderate gastritis groups.

Overall, the number of up-regulated genes seems to correlate with the degree of chronic inflammation. The 20 and 1 genes, respectively, that were selectively up- or down-regulated in the severe gastritis group might therefore relate to more severe chronic inflammation. The relative fold induction of the genes that were up-regulated either in the marked and/or moderated gastritis groups were also calculated and 10 of those showed a stronger and 1 showed a weaker expression in the severe gastritis group as compared with the moderate gastritis group (≥1.5-fold, Table V). These genes might therefore be quantitatively related to the intensity of the chronic inflammation.

Primers specific for H. pylori 16S rRNA, cagA, vacAs1/s2, and babA2 were applied in RT-PCR to detect the expression of the bacterial genes directly in the gastric biopsies. We could detect H. pylori 16S rRNA in all the samples belonging to the H. pylori-positive group, thus confirming the histological results (Table VI). cagA, vacAs1/s2, and babA2 were detected in selected H. pylori-positive samples (Table VI).

Table VI.

CagA, vacA, and babA2 genotyping by RT-PCR

Gene NamePatient Code
13911a1222274964658187
H. pylori 16S rRNA 
cagA − − − − − − 
vacAs1 − − − − − 
vacAs2 − − − − − − 
babA2 − − − − − − 
Gene NamePatient Code
13911a1222274964658187
H. pylori 16S rRNA 
cagA − − − − − − 
vacAs1 − − − − − 
vacAs2 − − − − − − 
babA2 − − − − − − 
a

Potentially infected by two different strains.

Four samples (1, 11, 12, and 64) tested positive for cagA/vacAs1/babA2 and they were also grouped together in the cluster analysis (Fig. 2,B, group I, right sub-branch). Samples 27, 49, 81, and 87 were positive only for the vacAs1 or vacAs2 genes and they were clustered together in group I, left sub-branch (Fig. 2 B). The genotype of the bacteria thus seems to correlate with the gene expression pattern to some extent. Therefore, we performed a multidimensional scaling to analyze the relationships among the samples infected by different bacterial strains. We found that all the H. pylori-negative samples were clustered together, as expected, suggesting a similar gene expression pattern. The H. pylori-positive samples were more different from each other, but no subgroup could be assigned, by any of the single or combined bacterial virulence factors.

We subsequently compared the gene expression in cagA+/cagA (n = 6/n = 6) or babA+/babA2− (n = 6/n = 6) samples. Thirty-five and 42 differentially expressed genes were identified (Mann-Whitney test, p < 0.05), and 19 and 25 of those showed >2-fold changes (supplementary data, Table IV). However, some of these genes showed rather low levels of significance (0.05 < p < 0.01), suggesting a limited biological relevance. One notable feature is that many of the H. pylori-infection signature genes were more prominent in biopsies taken from individuals infected by babA2-negative strains rather than positive strains (Fig. 2 B, marked by #). It is also notable that the differentially expressed genes in the cagA+/cagA or babA2+/babA2 groups are almost not overlapping, which suggests that different subsets of genes were induced by these bacterial factors. However, a larger sample size, which would allow more detailed comparisons (such as cagA+babA2/cagAbabA2 or babA2+cagA/babA2cagA), is needed to dissect the effects on gene expression by the individual bacterial factors.

To search for possible host factors that might contribute to the gene expression patterns, we performed series of statistical analyses, comparing samples from different gender and age groups (> and ≤ median age of 38). We could not identify any gene that was differentially expressed in males/females or old/young individuals using a high t threshold (p < 0.01). Using multidimensional scaling analysis, no subgroups could be assigned within the H. pylori-positive or -negative groups by age or gender. However, when we made comparisons of gender within the H. pylori-positive and -negative groups, we found 24 and 59 genes that were differentially expressed between males and females in the H. pylori-positive and -negative groups, respectively, although most showed a low level of significance (Mann-Whitney test, 0.01 < p < 0.05, data not shown). There were only three genes (CCR6, integrin αL, and transmembrane activator and CAML interactor) that showed a major difference in expression (Mann-Whitney test, p < 0.01), and they were up-regulated in females in the H. pylori-positive group. Similarly, we found that one gene, monocyte chemoattractant protein-2, was significantly up-regulated in older individuals (>38 years) within the H. pylori-positive group (Mann-Whitney test, p < 0.01). Therefore, gender and age are not influencing the general gene expression pattern in antrum mucosa in normal and H. pylori-infected individuals, but may contribute to differential expression of a small number of genes.

To the best of our knowledge, this study represents the first global assessment of expression of immunological/inflammatory genes in gastric biopsies in patients with H. pylori infection. Recently, several cDNA microarray analyses of H. pylori-mediated alterations of gene expression in gastric cancer or epithelial cell lines have appeared in literature (26, 27, 28, 29, 30). However, whether these systems properly reflect the situation in vivo is uncertain, particularly in view of the very limited overlap in the differentially expressed genes identified in our study and those published previously.

One notable finding of our study is that H. pylori infection strongly induced expression of several of genes that are associated with the innate immune system. Fourteen of the genes, identified as belonging to the H. pylori-infection signature, are related to innate immunity, including the recently described TLRs. Individual TLRs recognize distinct structural components of micro-organisms (31, 32). TLR4 recognizes LPS, the major outer membrane component of Gram-negative bacteria, whereas TLR2 is involved in the recognition of Gram-positive bacteria. The latter was the only TLR tested in our study (TLR1–6) that was not up-regulated in the H. pylori-infected samples, suggesting a selective expression pattern of TLRs in the infected gastric mucosa. Several other factors related to the LPS signaling pathway including CD14, LBP, and RP105 (related to proliferative response of B cells to LPS), were also strongly induced in the H. pylori-infected samples, suggesting that although LPS from H. pylori has been considered to be less toxic than LPS from other Gram-negative bacteria (33, 34, 35), it may still be a potent stimulator of innate immune responses. As gastric epithelial cells do not express TLR4 (36), H. pylori LPS is probably recognized by TLR4-expressing monocytes infiltrating the infected mucosa, as suggested previously in a mouse model (37). In addition to TLRs and related factors, expression of complement factor C3, LTF, a factor that mediates both antimicrobial and immunomodulatory activities by its iron-binding properties (38, 39, 40), and BPI were also strongly up-regulated in the H. pylori-infected gastric mucosa.

Up-regulation of the genes encoding receptors for IFN-γ, IL-2, IL-3, IL-12, and IL-18 was dominant in the H. pylori-infected gastric mucosa, supporting the previous notion of a Th1-driven response. The expression of IFN-γR1, previously not reported, was significantly up-regulated in H. pylori-infected mucosa. Interestingly, a recent genome-wide linkage analysis showed that specific polymorphisms in this gene are strongly associated with susceptibility to H. pylori infection (22). In addition, our study linked IL-10Rα, IL-15, IL-16, IL-18R, GM-CSFRβ, TNFSF3/LT-β and TNFSF7/CD27 gene expression to H. pylori infection.

Expression of selected chemokines and chemokine receptors was also modified during H. pylori infection. Among the 34 chemokine and 16 chemokine receptor genes tested, 7 and 3 genes, respectively, were identified as belonging to the H. pylori-infection signature. These included the CXC chemokines, ENA-78, GRO-γ, GRO-β, GRO-α, and BLC/BCA-1 (all target neutrophils except BLC/BCA-1 which targets naive B cells and activated T cells), which have previously been implicated in H. pylori-induced inflammation (41, 42, 43). In addition, two CC chemokines, pulmonary and activation-regulated chemokine (PARC) (DC-CK1) and MIP-3α, three chemokine receptors, CCR6, CCR7, and CXCR4, not previously studied in this regard, were also up-regulated. PARC is a dendritic cell-derived chemokine that preferentially attracts naive T cells (44). MIP-3α and its receptor CCR6 may play a role in attracting immature dendritic cells and lymphocytes within the memory subset (45) and increased MIP-3α production has been shown in epithelial cells isolated from patients with inflammatory bowel disease (46). MIP-3α may also mediate recruitment of memory CD4+ T lymphocytes through MadCAM-1. Interestingly, we found an up-regulation of the transcription of the MadCAM-1 gene in H. pylori-infected gastric mucosa, a finding also described in H. pylori-induced murine gastritis (47). Our data thus suggest that several CXC chemokines are involved in the neutrophil trafficking that contributes to the activity of H. pylori-induced gastritis, whereas PARC, MIP-3α, and BCA-1 may contribute to the recruitment of other inflammatory cells, including lymphocytes and dendritic cells.

The well-studied CXC chemokine IL-8 was expressed at a very low level (below the threshold for a genuine signal) in a majority of the biopsies tested and was therefore not analyzable by the array experiments. We could, however, detect an increased expression of IL-8 in the H. pylori-positive samples by real-time PCR (Mann-Whitney test, p < 0.005). However, the relative expression of this gene was very low, ∼20- to 200-fold less than in intestinal biopsies of patients with Crohn’s disease, and 200,000-fold less than in LPS-stimulated PBLs (S. Wen and Q. Pan-Hammarström, unpublished data). It is also worth noting that in contrast to IL-8, the expression of another CXC chemokine, ENA-78, was more dramatically increased, and quantitatively related to the degree of inflammation, which probably makes this gene a better marker for H. pylori-induced chronic inflammation. We further tested three additional genes that showed a very low level of expression in our array experiments, IL-2, IL-12p35, and IFN-γ by real-time PCR assay. There was a trend toward enhanced expression of these genes in H. pylori-positive samples (data not shown). However, the increment did not fulfill our stringent criteria for up-regulation and only a borderline significance was observed in these cases (Mann-Whitney test, p = 0.05, 0.03, and 0.02, respectively). Thus, array experiments may generate false negative results for genes with a low level of expression. Selected genes with a low level of expression, but with a proven biological relevance, should therefore be tested by real-time PCR. Alternatively, amplification of mRNA by template-switching PCR before array hybridization may allow identification of these genes (48).

Our cDNA array study also provided new insights on T cell responses to H. pylori infection. Two recently identified members of the B7 family of costimulatory factors, B7-H1 and GL50/B7-H2, were found to be strongly up-regulated in H. pylori-infected gastric mucosa. In contrast to the classical B7.1/B7.2-CD28 interactions, which are important for the priming naive T cells, these novel costimulatory interactions appear crucial in regulating effector lymphocytes in the periphery (49, 50). Inducible costimulator on activated T cells provides a positive signal by interaction with its ligand B7-H2, whereas programmed cell death-1 and its ligand B7-H1 provide negative signals to activated T cells that are crucial for the induction of peripheral tolerance and prevention of autoimmunity. A strong up-regulation of the two novel costimulatory factors in H. pylori-infected gastric mucosa samples suggests an important role for effector T cells at the sites of infection/inflammation regulated by both positive and negative signals/stimuli.

We also identified a number of genes (n = 21) that were associated with a more severe chronic inflammation induced by H. pylori infection. Some of these genes have already been implicated in the more severe clinical consequences of the infection. Examples are cyclin B1, a factor that determines cellular apoptosis (51) and previously associated with the progression of gastric mucosa-associated lymphoid tissue lymphoma (52) and MMP-1, an interstitial collagenase (53) shown to be related to H. pylori-induced ulcerogenesis (54, 55). Maspin expression was also up-regulated in the severe gastritis group. This protein, known as a protease inhibitor and a tumor suppressor gene (56), is expressed at a high level in gastric cells with intestinal metaplasia and may also contribute to gastric carcinogenesis (57). Further analysis of these genes may enable us to develop a set of molecular markers that can identify infected patients with a risk for developing a more severe disease.

Apart from the great advantage of array-based methods in analyzing gene expression in clinical samples, a number of pitfalls may also be encountered. First, the cellular sources of the gene expression changes are not identified and the relative contributions of epithelial cells and infiltrating leukocytes cannot readily be distinguished. Second, there are numerous regulatory events at the translational level that are not reflected by analysis of the transcriptional level as measured by array hybridization. Third, there may be false negative or positive results due to the sensitivity of the array assay and intrinsic limitations related to the design of the array. Other methods, such as quantitative real-time PCR and proteomic studies, are therefore needed to complement cDNA array experiments.

In summary, although the screened biopsies were limited in number, we found a remarkably consistent pattern of genes differentially expressed in normal and H. pylori-infected gastric mucosa. Numerous genes not previously associated with H. pylori were identified. The overall gene expression pattern in infected samples is probably associated with a number of factors, such as bacterial genotype, degree of inflammation and other host factors. Screening of a large number of biopsies from patients with different clinical phenotypes and/or infected by bacteria with different genotypes, and functional analysis of selected genes, might shed more light on the immune response against this pathogen.

Table V.

Gene expression in H. pylori-infected gastric mucosa with various degrees of chronic inflammationab

Gene NameGenbank Accession NumberDegree of Chronic GastritisCellular/Tissue Expressionc
Marked (n = 3)Moderate (n = 8)Mild (n = 1)
Adhesion molecules      
VCAM-1 NM_001078 ↑ ↑ ↑ Endo, D 
Contactin 1 NM_001843 ↑ Brain, pancreas, kidney 
MadCAM-1 NM_007164 ↑ ↑ Endo 
ICAM-1 NM_000201 ↑ Hemopoietic cells 
L-Selectin NM_000655 ↑ Leukocytes 
Apoptosis-related proteins      
ALOX5 NM_000698 ↑ ↑ Macro, Neutro, Mono, mast cells 
Binding proteins      
LBP NM_004139 ↑ ↑ Epi, hepatocytes 
BPI NM_001725 ↑ ↑ Epi 
Cell cycle regulators      
Cyclin B1d NM_031966 ↑ Overexpressed in malignant cells 
CDC25B NM_004358 ↑ Overexpressed in malignant cells 
Cell surface proteins      
CR1 NM_000573 ↑ Erythro, B, Mono, Neutro, Eosino, D 
TLR5 AF051151 ↑ ↑ Mono 
B7-H1 NM_014143 ↑ ↑ Endo, D 
CD53 NM_000560 ↑ ↑ B, Mono, Macro, Neutro, T 
TLR6 NM_006068 ↑ ↑ Mono, immature D, Endo 
SLAM NM_003037 ↑ ↑ Memory T, T, immature B 
C3 NM_000064 ↑ ↑ ↑ Mono, Macro 
CD45/Ly5 NM_002838 ↑ ↑ Hemopoietic cells 
TOSO NM_005449 ↑ ↑ ↑ 
RP105 NM_005582 ↑ ↑ ↑ 
TLR4 NM_003266 ↑ ↑ Mono, Macro, D, T 
CD21 (CR2) NM_001877 ↑ ↑ Mature B, T, D, Epi 
TLR1 NM_003263 ↑ ↑ Ubiquitous 
Chemokines and chemokine receptors      
GRO-γd NM_002090 ↑ ↑ Epi, Endo, Mono, Macro 
PARC NM_002988 ↑ ↑ Macro, D 
GRO-αd NM_001511 ↑ Epi, Endo, Mono, Macro 
BLC/BCA-1 NM_006419 ↑ ↑ 
ENA-78d NM_002994 ↑ ↑ Epi, Mono 
GRO-βd NM_002089 ↑ ↑ Epi, Endo, Mono, Macro 
MIP-3αd NM_004591 ↑ Endo, D, Mono, F 
Cytokines and cytokine receptors      
IFN-γR1 NM_000416 ↑ ↑ Macro, Mono, B, Endo 
PD-ECGF NM_001953 ↑ ↑ 
IFNAR2 NM_000874 ↑ ↑ Ubiquitous 
PF4 NM_002619 ↑ Megakaryocytes 
LIF Rd NM_002310 ↓ Muscle, prostate 
MER NM_006343 ↑ ↑ Neoplastic B and T cell lines 
FGF family members      
FGF-12 NM_021032 ↑ Brain 
Integrins      
α4 NM_000885 ↑ ↑ B, Thy, Mono, Granu, D 
αX NM_000887 ↑ ↑ Myeloid cells 
αL NM_002209 ↑ T, B, NK, Granu, Mono, Macro 
β2 NM_000211 ↑ ↑ Leukocytes 
ILs and IL receptors      
IL-16 NM_004513 ↑ T, mast cells, Eosino 
IL-2Rγ NM_000206 ↑ ↑ T, B, NK 
GM-CSFRβ NM_000395 ↑ ↑ Myeloid progenitors, Granu 
IL-1RL2 NM_003854 ↑ ↑ 
IL-10Rα NM_001558 ↑ ↑ B, Th, Mono, Macro, Epi 
IL-18RAP NM_003853 ↑ ↑ PBL 
Other factors      
NMA NM_012342 ↑ ↑ ↑ Kidney, placenta, spleen 
LTFd NM_002343 ↑ ↑ ↑ BM, spleen, lung 
GL50/B7-H2 AF199028 ↑ ↑ Epi, B, Mono, D 
(Table continues     
Gene NameGenbank Accession NumberDegree of Chronic GastritisCellular/Tissue Expressionc
Marked (n = 3)Moderate (n = 8)Mild (n = 1)
Adhesion molecules      
VCAM-1 NM_001078 ↑ ↑ ↑ Endo, D 
Contactin 1 NM_001843 ↑ Brain, pancreas, kidney 
MadCAM-1 NM_007164 ↑ ↑ Endo 
ICAM-1 NM_000201 ↑ Hemopoietic cells 
L-Selectin NM_000655 ↑ Leukocytes 
Apoptosis-related proteins      
ALOX5 NM_000698 ↑ ↑ Macro, Neutro, Mono, mast cells 
Binding proteins      
LBP NM_004139 ↑ ↑ Epi, hepatocytes 
BPI NM_001725 ↑ ↑ Epi 
Cell cycle regulators      
Cyclin B1d NM_031966 ↑ Overexpressed in malignant cells 
CDC25B NM_004358 ↑ Overexpressed in malignant cells 
Cell surface proteins      
CR1 NM_000573 ↑ Erythro, B, Mono, Neutro, Eosino, D 
TLR5 AF051151 ↑ ↑ Mono 
B7-H1 NM_014143 ↑ ↑ Endo, D 
CD53 NM_000560 ↑ ↑ B, Mono, Macro, Neutro, T 
TLR6 NM_006068 ↑ ↑ Mono, immature D, Endo 
SLAM NM_003037 ↑ ↑ Memory T, T, immature B 
C3 NM_000064 ↑ ↑ ↑ Mono, Macro 
CD45/Ly5 NM_002838 ↑ ↑ Hemopoietic cells 
TOSO NM_005449 ↑ ↑ ↑ 
RP105 NM_005582 ↑ ↑ ↑ 
TLR4 NM_003266 ↑ ↑ Mono, Macro, D, T 
CD21 (CR2) NM_001877 ↑ ↑ Mature B, T, D, Epi 
TLR1 NM_003263 ↑ ↑ Ubiquitous 
Chemokines and chemokine receptors      
GRO-γd NM_002090 ↑ ↑ Epi, Endo, Mono, Macro 
PARC NM_002988 ↑ ↑ Macro, D 
GRO-αd NM_001511 ↑ Epi, Endo, Mono, Macro 
BLC/BCA-1 NM_006419 ↑ ↑ 
ENA-78d NM_002994 ↑ ↑ Epi, Mono 
GRO-βd NM_002089 ↑ ↑ Epi, Endo, Mono, Macro 
MIP-3αd NM_004591 ↑ Endo, D, Mono, F 
Cytokines and cytokine receptors      
IFN-γR1 NM_000416 ↑ ↑ Macro, Mono, B, Endo 
PD-ECGF NM_001953 ↑ ↑ 
IFNAR2 NM_000874 ↑ ↑ Ubiquitous 
PF4 NM_002619 ↑ Megakaryocytes 
LIF Rd NM_002310 ↓ Muscle, prostate 
MER NM_006343 ↑ ↑ Neoplastic B and T cell lines 
FGF family members      
FGF-12 NM_021032 ↑ Brain 
Integrins      
α4 NM_000885 ↑ ↑ B, Thy, Mono, Granu, D 
αX NM_000887 ↑ ↑ Myeloid cells 
αL NM_002209 ↑ T, B, NK, Granu, Mono, Macro 
β2 NM_000211 ↑ ↑ Leukocytes 
ILs and IL receptors      
IL-16 NM_004513 ↑ T, mast cells, Eosino 
IL-2Rγ NM_000206 ↑ ↑ T, B, NK 
GM-CSFRβ NM_000395 ↑ ↑ Myeloid progenitors, Granu 
IL-1RL2 NM_003854 ↑ ↑ 
IL-10Rα NM_001558 ↑ ↑ B, Th, Mono, Macro, Epi 
IL-18RAP NM_003853 ↑ ↑ PBL 
Other factors      
NMA NM_012342 ↑ ↑ ↑ Kidney, placenta, spleen 
LTFd NM_002343 ↑ ↑ ↑ BM, spleen, lung 
GL50/B7-H2 AF199028 ↑ ↑ Epi, B, Mono, D 
(Table continues     
Table VA.

Continued

Gene NameGenbank Accession NumberDegree of Chronic GastritisCellular/Tissue Expressionc
Marked (n = 3)Moderate (n = 8)Mild (n = 1)
Proteases or related factors      
MMP-1d NM_002421 ↑ F, Macro 
BACE NM_012104 ↑ ↑ Brain, pancreas 
ADAM-19 NM_023038 ↑ Leukocytes 
ADAM12 NM_003474 ↑ ↑ Muscles, lung 
P15/Maspind NM_002639 ↑ Epi 
Signal transduction molecules      
TRAF1 NM_005658 ↑ Kidney, brain 
PTPN6 NM_002831 ↑ Hemopoietic cells 
CRAF1 NM_003300 ↑ 
PAK2 NM_002577 ↑ Ubiquitous 
PTPN7 NM_002832 ↑ Hemopoietic cells 
DRAK2 NM_004226 ↑ ↑ Spleen 
MAPK10d NM_002753 ↑ ↑ Brain, pancreas 
PRKCB1 NM_002738 ↑ ↑ ↑ Brain, BM, spleen 
TGF-β superfamily proteins      
AMHR2 NM_020547 ↑ ↑ Spleen 
TNF superfamily proteins      
EDAR NM_022336 ↑ ↑ Prostate, kidney, skin 
TNFRSF7 (CD27) NM_001242 ↑ ↑ Thy, T, NK, B 
TNFSF3 (LT-β) NM_002341 ↑ ↑ Mature T, B, NK 
TNFRSF1B NM_001066 ↑ T, B 
Total number of genes ↑ 66 47  
 ↓  
Gene NameGenbank Accession NumberDegree of Chronic GastritisCellular/Tissue Expressionc
Marked (n = 3)Moderate (n = 8)Mild (n = 1)
Proteases or related factors      
MMP-1d NM_002421 ↑ F, Macro 
BACE NM_012104 ↑ ↑ Brain, pancreas 
ADAM-19 NM_023038 ↑ Leukocytes 
ADAM12 NM_003474 ↑ ↑ Muscles, lung 
P15/Maspind NM_002639 ↑ Epi 
Signal transduction molecules      
TRAF1 NM_005658 ↑ Kidney, brain 
PTPN6 NM_002831 ↑ Hemopoietic cells 
CRAF1 NM_003300 ↑ 
PAK2 NM_002577 ↑ Ubiquitous 
PTPN7 NM_002832 ↑ Hemopoietic cells 
DRAK2 NM_004226 ↑ ↑ Spleen 
MAPK10d NM_002753 ↑ ↑ Brain, pancreas 
PRKCB1 NM_002738 ↑ ↑ ↑ Brain, BM, spleen 
TGF-β superfamily proteins      
AMHR2 NM_020547 ↑ ↑ Spleen 
TNF superfamily proteins      
EDAR NM_022336 ↑ ↑ Prostate, kidney, skin 
TNFRSF7 (CD27) NM_001242 ↑ ↑ Thy, T, NK, B 
TNFSF3 (LT-β) NM_002341 ↑ ↑ Mature T, B, NK 
TNFRSF1B NM_001066 ↑ T, B 
Total number of genes ↑ 66 47  
 ↓  
a

Relative fold induction = the average of the normalized gene expression of the selected gene in the severe gastritis group divided by the average of the normalized gene expression of the selected gene in the moderate gastritis group. FGF, fibroblast growth factor; PTPN, protein tyrosine phosphatase, nonreceptor; CRAF, CD40 receptor-associated factor; PAK, p21-activated kinase.

b

↑ and ↓, the gene expression is up- or down-regulated at least 2-fold as compared to H. pylori-negative samples with a normal histology (n = 5), or from nonexpression to expression and vice versa.

c

See footnote c for Table III.

d

Gene with ≥1.5 relative fold induction or reduction in the severe gastritis group as compared to the moderated gastritis group.

We thank Maribelle Herranz-Garcia for her excellent help in collecting the gastric biopsies.

1

This work was supported by the Swedish Foundation for International Cooperation in Research and Higher Education, the Swedish Doctors Association, the Swedish Society for Medical Research, Jonas Söderqvists Foundation, funds from the Karolinska Institute and Swiss National Foundation Grant 3100-068243 (to P.M.).

3

Abbreviations used in this paper: VacA, vacuolating cytotoxin; CagA, cytotoxin-associated Ag; BabA, blood group Ag-binding adhesin; TLR, Toll-like receptor; RAP, receptor accessory protein; RP105, radioprotective 105-kDa protein; CR, complement component receptor; BLC, B lymphocyte chemoattractant; BCA, B-lymphocyte chemoattractant; MIP, macrophage inflammatory protein; ENA, epithelia-derived neutrophil-activating peptide; GRO, growth-related protein; MAPK, mitogen-activated protein kinase; DRAK, DAP kinase-related apoptosis-inducing protein kinase; PRKCB, protein kinase C β; MMP, matrix metalloproteinase; ADAM, a disintegrin and metalloproteinase; SLAM, signal lymphocytic activation molecule; MadCAM, mucosal vascular addressin cell adhesion molecule; BPI, bactericidal/permeability increasing protein; LBP, LPS-binding protein; PD-ECGF, platelet-derived endothelial cell growth factor; LTF, lactotransferrin; AMHR, anti-Mullerian hormone receptor; PARC, pulmonary and activation-regulated chemokine; C3, complement factor 3; ALOX5, arachidonate 5-lipoxygenase; MER, c-mer proto-oncogene tyrosine kinase; EDAR, ectodysplasin 1 anhidrotic receptor.

4

The on-line version of this article contains supplemental material.

1
Parsonnet, J., G. D. Friedman, D. P. Vandersteen, Y. Chang, J. H. Vogelman, N. Orentreich, R. K. Sibley.
1991
. Helicobacter pylori infection and the risk of gastric carcinoma.
N. Engl. J. Med.
325
:
1127
.
2
The EUROGAST Study Group.
1993
. An international association between Helicobacter pylori infection and gastric cancer.
Lancet
341
:
1359
.
3
Xiang, Z., S. Censini, P. F. Bayeli, J. L. Telford, N. Figura, R. Rappuoli, A. Covacci.
1995
. Analysis of expression of CagA and VacA virulence factors in 43 strains of Helicobacter pylori reveals that clinical isolates can be divided into two major types and that CagA is not necessary for expression of the vacuolating cytotoxin.
Infect. Immun.
63
:
94
.
4
van Doorn, L. J., C. Figueiredo, R. Sanna, A. Plaisier, P. Schneeberger, W. de Boer, W. Quint.
1998
. Clinical relevance of the cagA, vacA, and iceA status of Helicobacter pylori.
Gastroenterology
115
:
58
.
5
Ilver, D., A. Arnqvist, J. Ogren, I. M. Frick, D. Kersulyte, E. T. Incecik, D. E. Berg, A. Covacci, L. Engstrand, T. Boren.
1998
. Helicobacter pylori adhesin binding fucosylated histo-blood group antigens revealed by retagging.
Science
279
:
373
.
6
Prinz, C., M. Schoniger, R. Rad, I. Becker, E. Keiditsch, S. Wagenpfeil, M. Classen, T. Rosch, W. Schepp, M. Gerhard.
2001
. Key importance of the Helicobacter pylori adherence factor blood group antigen binding adhesin during chronic gastric inflammation.
Cancer Res.
61
:
1903
.
7
Mahdavi, J., B. Sonden, M. Hurtig, F. O. Olfat, L. Forsberg, N. Roche, J. Angstrom, T. Larsson, S. Teneberg, K. A. Karlsson, et al
2002
. Helicobacter pylori SabA adhesin in persistent infection and chronic inflammation.
Science
297
:
573
.
8
Perez-Perez, G. I., B. M. Dworkin, J. E. Chodos, M. J. Blaser.
1988
. Campylobacter pylori antibodies in humans.
Ann. Intern. Med.
109
:
11
.
9
Crabtree, J. E., T. M. Shallcross, J. I. Wyatt, J. D. Taylor, R. V. Heatley, B. J. Rathbone, M. S. Losowsky.
1991
. Mucosal humoral immune response to Helicobacter pylori in patients with duodenitis.
Dig. Dis. Sci.
36
:
1266
.
10
Crabtree, J. E., J. D. Taylor, J. I. Wyatt, R. V. Heatley, T. M. Shallcross, D. S. Tompkins, B. J. Rathbone.
1991
. Mucosal IgA recognition of Helicobacter pylori 120 kDa protein, peptic ulceration, and gastric pathology.
Lancet
338
:
332
.
11
Tosi, M. F., S. J. Czinn.
1990
. Opsonic activity of specific human IgG against Helicobacter pylori.
J. Infect. Dis.
162
:
156
.
12
Negrini, R., L. Lisato, I. Zanella, L. Cavazzini, S. Gullini, V. Villanacci, C. Poiesi, A. Albertini, S. Ghielmi.
1991
. Helicobacter pylori infection induces antibodies cross-reacting with human gastric mucosa.
Gastroenterology
101
:
437
.
13
Negrini, R., A. Savio, B. J. Appelmelk.
1997
. Autoantibodies to gastric mucosa in Helicobacter pylori infection.
Helicobacter
2
:(Suppl.1):
S13
.
14
Negrini, R., A. Savio, C. Poiesi, B. J. Appelmelk, F. Buffoli, A. Paterlini, P. Cesari, M. Graffeo, D. Vaira, G. Franzin.
1996
. Antigenic mimicry between Helicobacter pylori and gastric mucosa in the pathogenesis of body atrophic gastritis.
Gastroenterology
111
:
655
.
15
Bamford, K. B., X. Fan, S. E. Crowe, J. F. Leary, W. K. Gourley, G. K. Luthra, E. G. Brooks, D. Y. Graham, V. E. Reyes, P. B. Ernst.
1998
. Lymphocytes in the human gastric mucosa during Helicobacter pylori have a T helper cell 1 phenotype.
Gastroenterology
114
:
482
.
16
D’Elios, M. M., M. Manghetti, M. De Carli, F. Costa, C. T. Baldari, D. Burroni, J. L. Telford, S. Romagnani, G. Del Prete.
1997
. T helper 1 effector cells specific for Helicobacter pylori in the gastric antrum of patients with peptic ulcer disease.
J. Immunol.
158
:
962
.
17
Sommer, F., G. Faller, P. Konturek, T. Kirchner, E. G. Hahn, J. Zeus, M. Rollinghoff, M. Lohoff.
1998
. Antrum- and corpus mucosa-infiltrating CD4+ lymphocytes in Helicobacter pylori gastritis display a Th1 phenotype.
Infect. Immun.
66
:
5543
.
18
Tomita, T., A. M. Jackson, N. Hida, M. Hayat, M. F. Dixon, T. Shimoyama, A. T. Axon, P. A. Robinson, J. E. Crabtree.
2001
. Expression of interleukin-18, a Th1 cytokine, in human gastric mucosa is increased in Helicobacter pylori infection.
J. Infect. Dis.
183
:
620
.
19
Wang, J., E. G. Brooks, K. B. Bamford, T. L. Denning, J. Pappo, P. B. Ernst.
2001
. Negative selection of T cells by Helicobacter pylori as a model for bacterial strain selection by immune evasion.
J. Immunol.
167
:
926
.
20
Suerbaum, S., P. Michetti.
2002
. Helicobacter pylori infection.
N. Engl. J. Med.
347
:
1175
.
21
El-Omar, E. M., M. Carrington, W. H. Chow, K. E. McColl, J. H. Bream, H. A. Young, J. Herrera, J. Lissowska, C. C. Yuan, N. Rothman, et al
2000
. Interleukin-1 polymorphisms associated with increased risk of gastric cancer.
Nature
404
:
398
.
22
Thye, T., G. D. Burchard, M. Nilius, B. Muller-Myhsok, R. D. Horstmann.
2003
. Genomewide linkage analysis identifies polymorphism in the human interferon-γ receptor affecting Helicobacter pylori infection.
Am. J. Hum. Genet.
72
:
448
.
23
Ibraghimov, A., J. Pappo.
2000
. The immune response against Helicobacter pylori-a direct linkage to the development of gastroduodenal disease.
Microbes Infect.
2
:
1073
.
24
Dixon, M. F., R. M. Genta, J. H. Yardley, P. Correa.
1996
. Classification and grading of gastritis: the updated Sydney system. International Workshop on the Histopathology of Gastritis, Houston 1994.
Am. J. Surg. Pathol.
20
:
1161
.
25
Eisen, M. B., P. T. Spellman, P. O. Brown, D. Botstein.
1998
. Cluster analysis and display of genome-wide expression patterns.
Proc. Natl. Acad. Sci. USA
95
:
14863
.
26
Chiou, C. C., C. C. Chan, D. L. Sheu, K. T. Chen, Y. S. Li, E. C. Chan.
2001
. Helicobacter pylori infection induced alteration of gene expression in human gastric cells.
Gut
48
:
598
.
27
Cox, J. M., C. L. Clayton, T. Tomita, D. M. Wallace, P. A. Robinson, J. E. Crabtree.
2001
. cDNA array analysis of cag pathogenicity island-associated Helicobacter pylori epithelial cell response genes.
Infect. Immun.
69
:
6970
.
28
Maeda, S., M. Otsuka, Y. Hirata, Y. Mitsuno, H. Yoshida, Y. Shiratori, Y. Masuho, M. Muramatsu, N. Seki, M. Omata.
2001
. cDNA microarray analysis of Helicobacter pylori-mediated alteration of gene expression in gastric cancer cells.
Biochem. Biophys. Res. Commun.
284
:
443
.
29
Bach, S., A. Makristathis, M. Rotter, A. M. Hirschl.
2002
. Gene expression profiling in AGS cells stimulated with Helicobacter pylori isogenic strains (cagA positive or cagA negative).
Infect. Immun.
70
:
988
.
30
Guillemin, K., N. R. Salama, L. S. Tompkins, S. Falkow.
2002
. Cag pathogenicity island-specific responses of gastric epithelial cells to Helicobacter pylori infection.
Proc. Natl. Acad. Sci. USA
99
:
15136
.
31
Akira, S..
2003
. Mammalian Toll-like receptors.
Curr. Opin. Immunol.
15
:
5
.
32
Takeda, K., T. Kaisho, S. Akira.
2003
. Toll-like receptors.
Annu. Rev. Immunol.
21
:
335
.
33
Asahi, M., T. Azuma, S. Ito, Y. Ito, H. Suto, Y. Nagai, M. Tsubokawa, Y. Tohyama, S. Maeda, M. Omata, et al
2000
. Helicobacter pylori CagA protein can be tyrosine phosphorylated in gastric epithelial cells.
J. Exp. Med.
191
:
593
.
34
Muotiala, A., I. M. Helander, L. Pyhala, T. U. Kosunen, A. P. Moran.
1992
. Low biological activity of Helicobacter pylori lipopolysaccharide.
Infect. Immun.
60
:
1714
.
35
Perez-Perez, G. I., V. L. Shepherd, J. D. Morrow, M. J. Blaser.
1995
. Activation of human THP-1 cells and rat bone marrow-derived macrophages by Helicobacter pylori lipopolysaccharide.
Infect. Immun.
63
:
1183
.
36
Bäckhed, F., B. Rokbi, E. Torstensson, Y. Zhao, C. Nilsson, D. Seguin, S. Normark, A. M. Buchan, A. Richter-Dahlfors.
2003
. Gastric mucosal recognition of Helicobacter pylori is independent of Toll-like receptor 4.
J. Infect. Dis.
187
:
829
.
37
Maeda, S., M. Akanuma, Y. Mitsuno, Y. Hirata, K. Ogura, H. Yoshida, Y. Shiratori, M. Omata.
2001
. Distinct mechanism of Helicobacter pylori-mediated NF-κB activation between gastric cancer cells and monocytic cells.
J. Biol. Chem.
276
:
44856
.
38
Aguila, A., A. G. Herrera, D. Morrison, B. Cosgrove, A. Perojo, I. Montesinos, J. Perez, G. Sierra, C. G. Gemmell, J. H. Brock.
2001
. Bacteriostatic activity of human lactoferrin against Staphylococcus aureus is a function of its iron-binding properties and is not influenced by antibiotic resistance.
FEMS Immunol. Med. Microbiol.
31
:
145
.
39
Guillen, C., I. B. McInnes, D. M. Vaughan, S. Kommajosyula, P. H. Van Berkel, B. P. Leung, A. Aguila, J. H. Brock.
2002
. Enhanced Th1 response to Staphylococcus aureus infection in human lactoferrin-transgenic mice.
J. Immunol.
168
:
3950
.
40
Weinberg, E. D..
2001
. Human lactoferrin: a novel therapeutic with broad spectrum potential.
J. Pharm. Pharmacol.
53
:
1303
.
41
Shimoyama, T., S. M. Everett, M. F. Dixon, A. T. Axon, J. E. Crabtree.
1998
. Chemokine mRNA expression in gastric mucosa is associated with Helicobacter pylori cagA positivity and severity of gastritis.
J. Clin. Pathol.
51
:
765
.
42
Mazzucchelli, L., A. Blaser, A. Kappeler, P. Scharli, J. A. Laissue, M. Baggiolini, M. Uguccioni.
1999
. BCA-1 is highly expressed in Helicobacter pylori-induced mucosa-associated lymphoid tissue and gastric lymphoma.
J. Clin. Invest.
104
:
R49
.
43
Innocenti, M., A. M. Svennerholm, M. Quiding-Jarbrink.
2001
. Helicobacter pylori lipopolysaccharides preferentially induce CXC chemokine production in human monocytes.
Infect. Immun.
69
:
3800
.
44
Adema, G. J., F. Hartgers, R. Verstraten, E. de Vries, G. Marland, S. Menon, J. Foster, Y. Xu, P. Nooyen, T. McClanahan, et al
1997
. A dendritic-cell-derived C-C chemokine that preferentially attracts naive T cells.
Nature
387
:
713
.
45
Liao, F., R. L. Rabin, C. S. Smith, G. Sharma, T. B. Nutman, J. M. Farber.
1999
. CC-chemokine receptor 6 is expressed on diverse memory subsets of T cells and determines responsiveness to macrophage inflammatory protein 3α.
J. Immunol.
162
:
186
.
46
Kwon, J. H., S. Keates, L. Bassani, L. F. Mayer, A. C. Keates.
2002
. Colonic epithelial cells are a major site of macrophage inflammatory protein 3α (MIP-3α) production in normal colon and inflammatory bowel disease.
Gut
51
:
818
.
47
Hatanaka, K., R. Hokari, K. Matsuzaki, S. Kato, A. Kawaguchi, S. Nagao, H. Suzuki, K. Miyazaki, E. Sekizuka, H. Nagata, et al
2002
. Increased expression of mucosal addressin cell adhesion molecule-1 (MAdCAM-1) and lymphocyte recruitment in murine gastritis induced by Helicobacter pylori.
Clin. Exp. Immunol.
130
:
183
.
48
Petalidis, L., S. Bhattacharyya, G. A. Morris, V. P. Collins, T. C. Freeman, P. A. Lyons.
2003
. Global amplification of mRNA by template-switching PCR: linearity and application to microarray analysis.
Nucleic Acids Res.
31
:
E142
.
49
Liang, L., W. C. Sha.
2002
. The right place at the right time: novel B7 family members regulate effector T cell responses.
Curr. Opin. Immunol.
14
:
384
.
50
Okazaki, T., Y. Iwai, T. Honjo.
2002
. New regulatory co-receptors: inducible co-stimulator and PD-1.
Curr. Opin. Immunol.
14
:
779
.
51
Porter, L. A., I. H. Cukier, J. M. Lee.
2003
. Nuclear localization of cyclin B1 regulates DNA damage-induced apoptosis.
Blood
101
:
1928
.
52
Banerjee, S. K., A. P. Weston, M. N. Zoubine, D. R. Campbell, R. Cherian.
2000
. Expression of cdc2 and cyclin B1 in Helicobacter pylori-associated gastric MALT and MALT lymphoma: relationship to cell death, proliferation, and transformation.
Am. J. Pathol.
156
:
217
.
53
Brinckerhoff, C. E., L. M. Matrisian.
2002
. Matrix metalloproteinases: a tail of a frog that became a prince.
Nat. Rev. Mol. Cell. Biol.
3
:
207
.
54
Menges, M., C. C. Chan, M. Zeitz, A. Stallmach.
2000
. Higher concentration of matrix-metalloproteinase 1 (interstitial collagenase) in H. pylori-compared to NSAID-induced gastric ulcers.
Z. Gastroenterol.
38
:
887
.
55
Yokoyama, T., Y. Otani, N. Kurihara, Y. Sakurai, K. Kameyama, H. Suzuki, N. Igarashi, M. Kimata, N. Wada, T. Kubota, et al
2000
. Matrix metalloproteinase expression in cultured human gastric wall fibroblasts-interactions with Helicobacter pylori isolated from patients with ulcers.
Aliment. Pharmacol. Ther.
14
:(Suppl. 1):
193
.
56
Bass, R., A. M. Fernandez, V. Ellis.
2002
. Maspin inhibits cell migration in the absence of protease inhibitory activity.
J. Biol. Chem.
277
:
46845
.
57
Son, H. J., T. S. Sohn, S. Y. Song, J. H. Lee, J. C. Rhee.
2002
. Maspin expression in human gastric adenocarcinoma.
Pathol. Int.
52
:
508
.
58
Atherton, J. C., P. Cao, R. M. Peek, Jr, M. K. Tummuru, M. J. Blaser, T. L. Cover.
1995
. Mosaicism in vacuolating cytotoxin alleles of Helicobacter pylori: association of specific vacA types with cytotoxin production and peptic ulceration.
J. Biol. Chem.
270
:
17771
.